# US 12,594,463 B2 · Method, device, and non-transitory computer-readable recording medium for estimating information on golf swing

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,594,463 B2
- Document ID: US12594463B2
- Status: Granted
- Kind: B2
- Application number: 18/033,694
- Filed: 2021-09-29
- Published: 2023-12-14
- Granted: 2026-04-07
- Earliest claimed: 2020-11-16
- Assignees: Moais, Inc.
- Applicants: Moais, Inc.
- Inventors: Yong Geun Lee
- Web page: https://patentgrep.com/patent/US12594463B2

## CPC

- A63B 24/0075: {Means for generating exercise programmes or schemes, e.g. computerized virtual trainer, e.g. using expert databases}
- A63B 71/0622: {Visual, audio or audio-visual systems for entertaining, instructing or motivating the user}
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/0495: Quantised networks; Sparse networks; Compressed networks
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06T 7/20: Analysis of motion (motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51)
- G06T 7/70: Determining position or orientation of objects or cameras (camera calibration G06T7/80)
- G06V 10/82: using neural networks
- G06V 40/23: {Recognition of whole body movements, e.g. for sport training}
- H04N 23/60: Control of cameras or camera modules
- A63B 2024/0012: {Comparing movements or motion sequences with a registered reference}
- A63B 2024/0068: {Comparison to target or threshold, previous performance or not real time comparison to other individuals}
- A63B 2071/063: {Spoken or verbal instructions}
- A63B 2102/32: Golf
- A63B 2220/05: Image processing for measuring physical parameters
- A63B 2220/806: Video cameras
- G06T 2207/10016: Video; Image sequence
- G06T 2207/20076: Probabilistic image processing
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/20132: Image cropping
- G06T 2207/30196: Human being; Person
- G06T 2207/30221: Sports video; Sports image
- A63B: APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT (apparatus for passive exercising, massage A61H)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

A method for estimating information on a golf swing is provided. The method includes the steps of: specifying at least one analysis target posture from among a plurality of golf swing-related postures; detecting at least one of at least one joint of a user and a golf club using an artificial neural network model from a video of the user's golf swing captured by a video capture module, and providing the user with information on the user's analysis target posture estimated with reference to at least one of a position of the at least one joint and a position of the golf club; and controlling the video capture module to capture the video of the user's golf swing again after a predetermined time elapses from when the information on the analysis target posture is provided.

## Claims

### Claim 1

A method performed in a mobile device for estimating information on a golf swing, the method comprising the steps of:
specifying at least one analysis target posture from among a plurality of golf swing-related postures;
light-weighting an artificial neural network model using depthwise convolution and pointwise convolution to obtain a light-weighted artificial neural network model, wherein a depth of each kernel in the depthwise convolution is reduced to 1 and each of a width and a height of each kernel in the pointwise convolution is reduced to 1;
detecting at least one of a golf club and at least one joint of a user using the light-weighted artificial neural network model from a video of a golf swing of the user captured by a video capture module, and providing the user with information on an analysis target posture of the user estimated with reference to at least one of a position of the at least one joint and a position of the golf club; and
controlling the video capture module to capture the video of the golf swing of the user again after a predetermined time elapses from when the information on the analysis target posture is provided.

### Claim 2 (depends on 1)

The method of claim 1, further comprising the step of:
when the capture of the video of the golf swing of the user is finished, providing the user with the information on the analysis target posture estimated while the video of the golf swing of the user is captured.

### Claim 3 (depends on 1)

The method of claim 1, wherein in the providing step, the user is further provided with a frame associated with the analysis target posture among a plurality of frames constituting the video.

### Claim 4 (depends on 1)

The method of claim 1, wherein the information on the analysis target posture includes a lesson content associated with the analysis target posture.

### Claim 5 (depends on 1)

The method of claim 1, wherein in the specifying step, the analysis target posture is specified with reference to information on a coach selected by the user from among a plurality of golf coaches.

### Claim 6

A non-transitory computer-readable recording medium having stored thereon a computer program for causing a mobile device for estimating information on a golf swing to:
specify at least one analysis target posture from among a plurality of golf swing-related postures;
light-weight an artificial neural network model using depthwise convolution and pointwise convolution to obtain a light-weighted artificial neural network model, wherein a depth of each kernel in the depthwise convolution is reduced to 1 and each of a width and a height of each kernel in the pointwise convolution is reduced to 1;
detect at least one of a golf club and at least one joint of a user using the light-weighted artificial neural network model from a video of a golf swing of the user captured by a video capture module, and provide the user with information on an analysis target posture of the user estimated with reference to at least one of a position of the at least one joint and a position of the golf club; and
control the video capture module to capture the video of the golf swing of the user again after a predetermined time elapses from when the information on the analysis target posture is provided.

### Claim 7

A mobile device for estimating information on a golf swing, the mobile device comprising an application configured to:
specify at least one analysis target posture from among a plurality of golf swing-related postures;
light-weight an artificial neural network model using depthwise convolution and pointwise convolution to obtain a light-weighted artificial neural network model, wherein a depth of each kernel in the depthwise convolution is reduced to 1 and each of a width and a height of each kernel in the pointwise convolution is reduced to 1;
detect at least one of a golf club and at least one joint of a user using the light-weighted artificial neural network model from a video of a golf swing of the user captured by a video capture module, and provide the user with information on an analysis target posture of the user estimated with reference to at least one of a position of the at least one joint and a position of the golf club; and
control the video capture module to capture the video of the golf swing of the user again after a predetermined time elapses from when the information on the analysis target posture is provided.

### Claim 8 (depends on 7)

The mobile device of claim 7, wherein the application is configured to, when the capture of the video of the golf swing of the user is finished, further provide the user with the information on the analysis target posture estimated while the video of the golf swing of the user is captured.

### Claim 9 (depends on 7)

The mobile device of claim 7, wherein the application is configured to further provide the user with a frame associated with the analysis target posture among a plurality of frames constituting the video.

### Claim 10 (depends on 7)

The mobile device of claim 7, wherein the information on the analysis target posture includes a lesson content associated with the analysis target posture.

### Claim 11 (depends on 7)

The mobile device of claim 7, wherein the application is configured to specify the analysis target posture with reference to information on a coach selected by the user from among a plurality of golf coaches.